Traditional metal-oxide-semiconductor (MOS) transistors often use metal silicide layers to reduce resistance. A self-aligned silicidation process (salicide) is often used to form the region of titanium, cobalt or tungsten silicide on the gate electrode and source/drain regions of the MOS transistor. In this process, a blanket metal film is deposited on the silicon substrate containing the MOS transistor structure. The metal is then reacted with the underlying silicon regions to form a low resistance metal silicide. Any unreacted metal remaining on the substrate is then removed using a metal etch process that is selective to the remaining metal silicide.
In order to reduce the resistances associated with the metal silicide regions, nickel is finding increasing use in forming the metal silicide regions in MOS transistors, particularly for transistors with physical gate lengths of less than 40 nm and/or MOS transistors with ultra-shallow junctions. Nickel, unfortunately, has a very high diffusivity in silicon leading to the formation of nickel silicide regions that extend beneath the transistor sidewall structures. Regrettably, the nickel silicide regions that extend beneath the transistor sidewall structures tend to lead to nickel silicide excessive encroachment defects extending into the channel region of the MOS transistor. In a similar manner, the high diffusivity of nickel causes excessive spike defects into the source/drain regions. As would be expected, the encroachment and spike defects tend to cause serious acceptance, manufacturability, and ultimately device yield problems.
There is therefore a need for a method to form metal silicide regions in MOS transistors that does not experience the severe defect issues caused by the traditional methods.